ReleaseMicrosoftMicrosoftpublished Aug 28, 2024seen 5d

microsoft/onnxruntime-extensions v0.12.0

microsoft/onnxruntime-extensions

Open original ↗

Captured source

source ↗
published Aug 28, 2024seen 5dcaptured 8hhttp 200method plain

v0.12.0

Repository: microsoft/onnxruntime-extensions

Tag: v0.12.0

Published: 2024-08-28T17:16:45Z

Prerelease: no

Release notes:

What's Changed

  • Added C APIs for language, vision and audio processors including new FeatureExtractor for Whisper model
  • Support for Phi-3 Small Tokenizer and new OpenAI tiktoken format for fast loading of BPE tokenizers
  • Added new CUDA custom operators such as MulSigmoid, Transpose2DCast, ReplaceZero, AddSharedInput and MulSharedInput
  • Enhanced Custom Op Lite API on GPU and fused kernels for DORT
  • Bug fixes, including null bos_token for Qwen2 tokenizer and SentencePiece converted FastTokenizer issue on non-ASCII characters, as well as necessary updates for MSVC 19.40 and numpy 2.0 release

New Contributors

  • @yihonglyu made their first contribution in https://github.com/microsoft/onnxruntime-extensions/pull/702
  • @skyline75489 made their first contribution in https://github.com/microsoft/onnxruntime-extensions/pull/748

Full Changelog: https://github.com/microsoft/onnxruntime-extensions/compare/v.0.11.0...v0.12.0

Notability

notability 3.0/10

Routine library release, low traction